Acute Diseases: Pneumonia and Influenza
Obtaining figures on the incidence and burden of pneumonia and influenza in adults throughout the developing world has been surprisingly difficult. Much of the research and surveillance has been directed toward the pediatric age group (see chapter 25). In 2000, fatal lower respiratory infections, as a class that represents serious pneumonia and influenza, were reported as the cause of 120 deaths per million men and 76 deaths per million women worldwide for the 15 to 59 age group (WHO 2000). For both sexes in this age group, this statistic represents approximately one-third of the deaths caused by tuberculosis. However, for the age groups over 60, rates of death from lower respiratory disease more than double for each decade of life, whereas rates of death from tuberculosis remain relatively constant. Notably, acute respiratory diseases—in addition to tuberculosis—remain major concerns among adults with AIDS.
The diagnosis of pneumonia varies according to the patient's access to medical care. Often the diagnosis is made simply on the basis of cough and fever. For patients with access to a hospital, the likelihood of obtaining a chest x-ray increases; generally the infection is bacteriologically confirmed only in the most sophisticated medical centers. The natural history of pneumonia without antibiotic treatment varies with the etiologic agent and the patient's underlying comorbid conditions and age. Mortality resulting from these lower respiratory diseases is approximately 10-fold higher in people age 60 to 69 than in people age 15 to 59 (WHO 2000). Comorbid conditions, malnutrition, low socioeconomic status, and cigarette smoking each play a role in increasing the incidence of disease and worsening the prognosis, both with and without treatment.
From studies conducted in the developed world, it would be reasonable to conclude that common antibiotics for pneumonias that occur outside a hospital setting would effectively reduce days lost from work and, in the absence of other morbid conditions, mortality. The few studies in which sputum specimens have been cultured suggest that Streptococcus pneumoniae is found in between 40 and 50 percent of the cases. Gram-negative organisms or mixed infections are often isolated, and thus, the use of broad-spectrum antibiotics is warranted (Hooi, Looi, and Ng 2001; Hui and others 1993; Lieberman and others 1996). As would be expected, increased use of antibiotics has resulted in increased resistance to common antibiotics. In addition, 10 to 15 percent of these cases may be tuberculosis (Dolin, Raviglione, and Kochi 1994).
Scott and others (2000) suggested that, despite the similarity of the mortality rates for hospital-treated pneumonia in developing and developed countries, there are important differences in the age distributions. The median age at death among Kenyan adults was 33 years, in contrast to more than 65 years in more developed countries. Many patients in developing countries present late in the course of the disease. Often they die before an appropriate diagnostic workup can be completed, thus leading to an underestimate of case-fatality rates.
Signs and symptoms of influenza can vary from trivial to explosive. Although the disease is usually self-limiting, it can result in both severe incapacity and, when not properly treated, potentially fatal secondary pneumonia. Clearly, patients with comorbid conditions, the very young, pregnant women, and the elderly are at greater risk of suffering from complications from influenza. Those criteria, along with the adequacy of supply, form the basis for choosing who should be considered for vaccination each year. Because the symptoms of influenza can be quite similar to those of bacterial pneumonia, influenza may often be misdiagnosed as pneumonia. Generally, influenza is more self-limiting than pneumonia, although the infectivity and transmission of influenza from person to person can be substantial. The current threat of H5N1 influenza has resulted in increased human and avian surveillance and preparations for a possible pandemic (box 35.1).
[Box 35.1]
The recent 2003 outbreak of severe acute respiratory syndrome (SARS; see chapter 53) emphasizes the importance of accurate and open surveillance and a coordinated response in controlling the spread of newly active influenza strains. The potential for global spread and the occurrence of worldwide epidemics of influenza (presumed to be transmitted to humans from domesticated or wild animals and then through close proximity to humans with symptomatic disease—generally to caregivers) points out the importance of continued surveillance for such episodes (Low and McGeer 2004). The lessons learned from the SARS epidemic reinforce the importance of proven traditional public health measures, such as finding and isolating cases, quarantining close contacts, and improving infection control practices (Bell 2004). Those methods, along with several other, less traditional efforts, were presumed to be part of the reason the epidemic was contained as promptly as it was (see table 35.1). However, because of the high case-fatality rate, the disease caused significant disruption throughout the world.
[Table .]
Economic Impact of Influenza and Cost-Effectiveness of Interventions in the Developed World
Influenza is common in developed countries. Annually, it affects 10 to 20 percent of the U.S. population (Lee and others 2002); those affected experience on average a loss of 2.8 workdays per episode. Those over 65 years of age are more susceptible to complications, increased costs of hospitalization, and even death. The cost of outbreaks can be large. The costs of the 1996-97 epidemic in Germany were estimated at US$1,045 million, and the annual costs of outbreaks at US$11 million to US$18 million (WHO 2002a).
For those over age 65, many countries encourage preventive vaccinations annually, on the basis of studies suggesting that vaccination (either opportunistic or in a campaign) is cost-effective in elderly populations (for example, see the model of Scuffham and West 2002). Given a good antigenic match, inactivated influenza vaccines prevent laboratory-confirmed illness in 70 to 90 percent of healthy adult vaccine recipients (WHO 2002a). Vaccination is less costly than chemoprophylaxis (with ion-channel inhibitors such as amantadine and rimantadine, or with neuraminidase inhibitors such as zanamivir and oseltamivir) or early treatment with the same drugs. In both the institutionalized and the healthy elderly, vaccination substantially reduces overall mortality from influenza (by 40 to 68 percent).
The cost-effectiveness of vaccination for healthy working-age adults, taking into account workdays lost, is a matter of debate. Demicheli and others (2000) concluded that the most cost-effective option for healthy adults age 14 to 60 was to take no action. However, these authors include only medical costs in their calculations. Postma and others (2002) reviewed 11 studies. Only one shows cost savings on the basis of medical costs alone, but nine of them implied cost savings from vaccination if the value of lost work is included. Because of differences in costs and health care usage patterns, data on cost savings in developed countries cannot be helpfully extrapolated to developing countries.
Economic Impact of Influenza in the Developing World
In Hong Kong, China (where there is a milder year-round pattern of infection, little influenza-related mortality, and low reported work losses), a model suggested that vaccination was not cost saving, even if targeted to the elderly (Fitzner and others 2001). The only case for vaccination was if it controlled the emergence of highly virulent strains and prevented transmission to the rest of the world. According to the World Health Organization (WHO), much less is known about the impact of influenza in the developing world. However, in the tropics, where viral transmission normally continues year-round, influenza outbreaks tend to have high attack and case-fatality rates. For example, during an influenza outbreak in Madagascar in 2002, more than 27,000 cases were reported within three months and 800 deaths occurred despite rapid intervention. An investigation of this outbreak, coordinated by WHO, found that health consequences were severe in poorly nourished populations with limited access to adequate health care (WHO 2002b). It is not possible to extrapolate the exact annual burden of influenza in the tropics from data on such occasional and severe outbreaks. Because many areas (for example, Sub-Saharan Africa) do not have surveillance centers, not enough is known at this point to make policy recommendations. There are also no readily available estimates of the cost-effectiveness of influenza vaccination in those environments. (For further discussion of the role of vaccination, see chapter 20.)
